Accuracy and time efficiency of deep learning-based method for single-crown design compared with a conventional CAD...

Area of Science:

• Dental Technology
• Artificial Intelligence in Dentistry
• Digital Dentistry

Background:

• Designing dental restoration contours is traditionally labor-intensive.
• Computer-aided design and computer-aided manufacturing (CAD-CAM) systems automate aspects of crown design.
• The comparative accuracy of deep learning (DL) versus conventional CAD for single-tooth crowns is not well-established.

Purpose of the Study:

• To systematically review and compare the accuracy and time-efficiency of DL-based software versus conventional CAD systems for single-tooth crown design.
• To evaluate the performance of DL in reproducing occlusal morphology and finish line details.

Main Methods:

• A systematic review adhering to PRISMA guidelines was conducted.
• Searches across 5 databases up to July 2025 identified in vitro studies.
• Studies compared DL software with conventional CAD for single-tooth crown reconstruction and worktime; methodological quality was assessed using MINORS.

Main Results:

• Seven studies met inclusion criteria.
• DL-based software demonstrated superior or comparable accuracy in contour design and finish line detection versus conventional CAD.
• Internal fit discrepancies were clinically acceptable (55.4–83.1 µm); DL significantly reduced working time (P<.05).

Conclusions:

• DL-based software offers enhanced accuracy for occlusal morphology and finish line design in single-tooth crowns.
• DL systems significantly decrease the time required for crown design compared to conventional CAD.
• DL represents a promising advancement for efficient and accurate digital dentistry workflows.